Powering 100,000+ developers and the world's most forward-thinking companies

Watch your virtual agent come to life

PALs deliver the most natural real-time, conversational video AI experience, powered by multimodal research models.

real-time conversational AI

Talk to a PAL

FACIAL RENDERING

It looks like a human.

Lifelike rendering with real expression and micro-motion.

VOICE & CONVERSATION

It knows when to speak.

Expressive, human-like speech that flows like a natural conversation.

AUDIO-VISUAL PERCEPTION

It sees and hears you.

Your PAL sees what is in frame and reacts to the context.

MEMORY & INTELLIGENCE

It remembers you.

Every prior conversation carries forward, so trust compounds.

Bespoke PALs for every use case

Customers big and small are building PALs to scale their organization, drive revenue, and build trust and connection with users.

SALES ASSISTANT

A go-to-market agent that lives on your website or joins sales calls live to answer product questions, engage buyers, and drive conversion.

INTERVIEWER

Human-like AI interviewers that screen thousands of candidates at once, helping you find the best talent without adding recruiter hours.

HEALTHCARE

HIPAA-compliant agents built for patient intake, support, education, and other high-touch healthcare experiences.

LEARNING & DEVELOPMENT

Realistic AI simulations for training, coaching, and practice for an engaging and scalable learning environment.

BUILD YOUR OWN

Build a conversational AI agent for any use case. Give it a face, voice, knowledge, and instructions, then deploy it anywhere your users need it.

1

1

6

1

Visual reasoning

Emotion understanding

Perception tools

Visual reasoning

Emotion understanding

Perception tools

Visual reasoning

Emotion understanding

Perception tools

Visual reasoning

Emotion Understanding

Perception tools

AI that can work like humans do

Giving your PALs all the tools to make for interactive experiences.

FACE-TO-FACE VIDEO

Presentation mode

Present a slide deck, share slides, and talk through them on their own.

Browser Mode

Open your product, share their screen, click through it, and guide someone through the experience while talking with them live.

Knowledge Base

Upload your document and your PAL answers straight from them, grounded in your facts, not guesses.

Magic Canvas

Generate interactive visuals like charts, questionnaires, or scheduling pages, on the fly during a live video conversation.

Google Meet and Zoom

Join a Google Meet as a full participant, on camera, seeing and hearing the room, and taking its turn in real time.

Deploy your PAL today

Build it with our no-code platform or API, or let us handle the whole thing.

PAL MAKER

Creating with Tavus PAL Maker

For Builders: PAL Maker

PALs are now easier to make than ever. Create and launch a PAL in minutes with the PAL Maker, no code or engineering required.

Get Started Now

DEVELOPER API

Tavus Code API Graphic

For Developers: API Platform

Build advanced and deeply customized PAL experiences with APIs for real-time, face-to-face AI conversations.

Explore API

Custom SOLUTIONS

For Enterprises: Solutions

Solutions are bespoke, fully managed PAL deployments. Our team designs, builds, tunes, and scales your PAL from concept to production, tailored to your business and built to perform.

contact salEs
Our Vision

The human computing company

Tavus is an AI research lab pioneering human computing: a future where using a computer feels as natural as talking to a friend or co-worker. We build foundational models that teach machines to see, hear, understand, and respond, making AI feel unmistakably human.

See Careers
read research

models

Four models, working in sync.
One seamless conversation.

Tavus unifies turn-taking, rendering, perception, and speech in real time, so every interaction feels immediate, aware, and natural.
Our Research
White background with light blue dotted decorative swirls and curls on edges
Best-in-class TURN-TAKING

Sparrow [2]

Sparrow is the #1 conversational understanding model on the market, built to understand not just what someone says, but when to listen, wait, speak, and respond. It understands the entire audio scene - including pauses, interruptions, backchannels, multiple speakers, and background noise - to make AI conversations feel more naturally human.
4× faster inference - processes 80ms of audio in ~7ms.
92.4% end-of-turn recall - outperforming every other published system tested.
Best-in-class rendering

Phoenix [4.5]

Phoenix is the most natural and realistic human rendering model on the market, designed to generate lifelike faces, expressions, gestures, and upper-body movement in real time.  It can render entirely new identities zero-shot from a single image, making photorealistic conversational AI faster and easier to create than ever before.
134ms audio-to-video latency
25% faster than any other model on the market
Best-in-class perception

Raven [1]

Raven is a new type of multi-modal perception model, giving machines the ability to see, hear and understand in real-time. It translates facial expressions, tone, gaze, emotion, and environmental context into rich conversational signals, helping your AI respond with empathy, awareness, and intent.
<100ms audio perception
<300ms context freshness - continuously updating its understanding of your emotional and attentional state
Coming soon

Coming Soon

Unlocking AI for everyone

Start with a PAL of your own, or bring us the problem and we will build it with you.

Build a pal
Talk to Sales
Soft black oval shadow with blurred edges on a transparent background.

THE HUMAN COMPUTING MOMENT

Pixel art of large orange clouds against a clear blue sky.A detailed painting of a left hand with fingers slightly bent, palm facing down.Pixelated right hand pointing with index finger extended to the left.
Close-up of a flexed muscular left arm with a visible bicep.